Get current baton for a pipeline.
AI agents call get_relay_baton to retrieve information from Agent Runtime without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves the current state of a relay baton (likely a control token or checkpoint in a multi-agent pipeline) without modifying, executing, deleting, or creating anything. It is a pure read operation with minimal blast radius if misused by an AI agent.
From the tool's definition Tool name 'get_relay_baton' and description 'Get current baton for a pipeline' indicate data retrieval only. The verb 'get' and absence of modification language confirm no side effects.
Attacks that exploit this kind of access
Get current baton for a pipeline. It is categorised as a Read tool in the Agent Runtime MCP Server, which means it retrieves data without modifying state.
Register the Agent Runtime MCP server in PolicyLayer and add a rule for get_relay_baton: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Agent Runtime. Nothing to install.
get_relay_baton is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_relay_baton rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for get_relay_baton. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
get_relay_baton is provided by the Agent Runtime MCP server (marc-shade/agent-runtime-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →